Integrative machine learning reveals interplay between epithelial and endothelial cells elevates mixed heavy metal-related colorectal cancer risk.

Heavy metal contamination has become an emerging global health concern, yet the promoting effects of mixed heavy metal exposure on colorectal cancer risk remains poorly understood. In this study, plasma concentrations of 25 heavy metals were obtained from an in-house elemental database comprising both colorectal cancer patients and healthy controls, and exposure concentrations were then defined based on these human plasma detection data and relevant public datasets. Machine learning algorithms and mixture exposure models were applied to identify key heavy metals associated with colorectal cancer risk. RNA-seq datasets were integrated to screen for genes exhibiting heavy metal-tumor-specific characteristics. A total of 270379 cells from 95 samples were analyzed to identify cell subsets sensitive to mixed heavy metal exposure. Molecular and cellular experiments were performed to validate the underlying mechanisms. We first identified stannum (Sn), antimony (Sb), tantalum (Ta), and thallium (Tl) as a mixture of heavy metal highly associated with colorectal cancer risk. ESM1, SLC7A5 and GRIA4 were further identified as key genes linked to colorectal tumors under conditions of mixed heavy metal exposure. Based on scRNA-seq analyses, endothelial tip cells were first identified as exhibiting heightened sensitivity to mixed heavy metal exposure. Mechanistically, acute mixed heavy metal exposure upregulated VEGFA expression in colorectal tumor epithelial cells. Co-culture experiments demonstrated enhanced epithelial-endothelial communication through the VEGFA-FLT1 axis, accompanied by increased expression of ESM1 and ANGPT2 in endothelial cells. Collectively, these findings suggest that mixed heavy metal exposure promotes angiogenesis by enhancing the interaction between colorectal tumor epithelial and endothelial cells, thereby increasing the risk of colorectal cancer. This study provides a scientific basis for improving strategies to control heavy metal contamination.